The long range objective is to provide a better understanding of how regulatory proteins interact with each other and with DNA to control gene expression in procaryotes. The four operons in Escherichia coli involved in the assimilation of L-arabinose will be studied. Three specific questions will be addressed. (1) How is transcription of the araBAD, araE and araFG operons activated by both AraC protein and cyclic-AMP receptor protein? (2) How does AraC protein repress transcription of the operons when bound at DNA sites well upstream of the promoter regions required for induction? (3) What are the specific protein and DNA elements involved in AraC recognition of DNA binding sites? The locations of specific contact points between DNA polymerase and promoter DNA will be determined in vitro by DNA chemical premodification techniques using the gel electrophoresis DNA-binding assay. Previously undetected promoter sequences required only during intermediate stages of the binding process also can be revealed with this technique. Site-directed mutagenesis with oligonucleotides will be used to test the function in vivo of specific contact sites. Kinetic measurements will be made for RNA polymerase binding to each of the promoters using the abortive initiation assay as well using the more direct gel electrophoresis-DNA binding assay. With these data the individual contributions of AraC and cycli c-AMP receptor protein in stimulating promoter binding can be assessed in terms of the initial binding and the rate of isomerization to the active "open complex." Promoter structure-function relationships can be studied by correlating binding parameters with differences among the promoters in the relative positions of the activators. Mutants in the regulatory proteins and in the promoter sites will be used for studies in vitro and in vivo. AraC mutants defective in activating particular promoters will be especially useful in studying protein-protein interactions and in studying the mechanism of repression from upstream DNA sites. AraC protein binds two apparent classes of DNA sites that can be distinguished by the extent of the protein contact region along the DNA. AraC mutants will be selected that bind one, but fail to bind another site. These and additional mutations in AraC and in the DNA binding sites will aid in determining the protein domains involved in binding. The mutants should also provide information on the symmetry of the DNA sites.